Summary
we report in this review the studies on the regulation of cancer growth by the factors present in embryo during precise stages of cell differentiation. These report summarizes the in vitro and in vivo results of a series of experimental evidence gathered during the last 18 years, covering several approaches such as tests on animal model, citotoxicity assays, immunohistochemical, flow cytometry and molecular biology methods. These studies on the normal cell differentiation and on the regulation of the pathological growth of mutated undifferentiated cells such as cancer cells, have highlighted the way about the use of the embryonic stem cells differentiation stage factors in the treatment of tumors. The studies about the use of stem cells differentiation stage factors in controlling tumor growth are in progress and they will be the subject of the next manuscript.
Introduction
During pregnancy, a close cross-talk between mother and developing embryo is formed, made of a complex network of circulating molecular factors. This cross-talk is necessary for the prevention of pregnancy-threatening events, including the establishment of abnormally proliferating cell clones which may damage the integrity of this cross-talk. This problem had already been encountered by other groups, which pointed out at activated T-cell clones across mother-embryo interface.1,2 In previous studies a model which links embryonic development to carcinogenesis, i.e., the onset of an abnormous proliferative program, was described.3 The development of a theoretical model of gene regulation of tumor cells by factors of embryonic microenvironment comes after a several years-long work, which involved in vitro and in vivo approaches. In this short review, the results of this work are summarized.
In vitro results
Administration of substances of zebrafish embryos taken during precise stages of development, i.e., 1000 blastomeres (a full cleavage stage), 50% epiboly (corresponding to the onset of gastrulation), 5 somites and 20 somites was performed on glioblastoma, melanoma, kidney adenocarcinoma, breast carcinoma and lymphoblastic leukemia cells, and proliferation curves were drawn 24 and 48 hours after the treatment.4 All cell lines exhibited a slowing down of their proliferation values when treated with substances taken during the stages of cell differentiation, i.e., from 50% epiboly. (Fig. 1,2,3,4,5) Each cell line followed a peculiar rate of slowing and showed a specific response to treatment with the different developmental stages: for example, the proliferation of glioblastoma was inhibited most by the 50% epiboly stage and least by the 20 somites stage, whereas melanoma cells were slowed most by the 5 somites stage and kidney adenocarcinoma cells by 20 somites stage. Glioblastoma, melanoma, breast carcinoma and lymphoblastic leukemia cells significantly responded to treatment up from 24 hours, whereas kidney adenocarcinoma cells did not exhibit any slowing at all. At the 48 hours endpoint, however, all cell lines were affected by the treatment, with inhibition percent values ranging from 73% of glioblastoma cells and 26% of melanoma cells treated with substances taken from epiboly stage ( during which the embryonic totipotent stem cells are differentiating into the pluripotent stem cells).
No slowing response was shown by cells treated with substances taken from a developmental stage prior to 50% epiboly, namely, the 1000 blastomeres stage (called stage 1k).
On the contrary, treated cells exhibited a weak proliferative response. (Fig. 6,7,8,9) This evidence enforces our view that differentiative stages of development are characterized by regulatory networks which re-direct tumor cells to a normalized path of differentiation, and that these networks appear from the onset of gastrulation. Before gastrulation, embryos are subjected to merely proliferative stimulating networks which fail to normalize tumor cells, possibly enhancing their abnormal growing potential.
A similar slowing effect on cell growth was observed after the administration of extracts of crude pregnant uterine mucosa to the same tumor cell lines.5 Cells treated with uterine mucosa substances taken from 23 days-pregnant sow exhibited slower proliferation curves, with inhibition percent values ranging from 80% of breast carcinoma cells and 67% of lymphoblastic leukemia cells to 22% of glioblastoma cells. As well as previously observed with substances taken from zebrafish embryo, each cell line responded to treatment with uterine mucosa extracts in a peculiar manner. The day of pregnancy at which the mucosa was collected seems not to affect the response of tumor cells, since the treatment of glioblastoma cells with uterine mucosa substances taken from different days-pregnant mice led to a diffuse slowing down of the proliferation curve, with inhibition percent values not significantly differing from each other. Finally, the effect of uterine mucosa substances on cell growth appears to be ascribed to tumour cell lines only, since the treatment of a non-tumoral line, murine fibroblast NIH 3T3, did not lead to a change of the cell proliferation rate. (Fig. 10)
In order to elucidate which factors are responsible of this effect, we fractionated the whole uterine extract from pregnant pig by low-molecular weight cutoffs, and finally we isolated a 5 kDa fraction which retained the slowing-down efficacy on tumor cell growth. (Fig. 11) We called this fraction “Life-Protecting Factor” (LPF), because it may contain the molecular factors involved in preserving embryo-mother integrity from pathological cell clones. It is likely that the mechanism of action is apoptosis-mediated, since we observed high levels of a nucleosomal fraction in the medium of tumor cells after 24 hours-treatment with uterine substances . (Fig. 12)
Possible molecular bases of this proliferation-slowing mechanism on cell lines were also investigated by several techniques. Flow cytometry analysis revealed a mean 20% increase of the expression of tumor suppressor p53 in glioblastoma and melanoma cells after the treatment with substances taken from Zebrafish embryo.6 (Fig. 13,14) Immunohistochemical analysis on treated melanoma and hepatocarcinoma cells showed a dramatic increase of p53 staining respect to untreated cells.6 It has to be noted that not all embryonic substances were able to induce p53 overexpression, confirming that only precise differentiation stages bear the tumor growth-slowing potential. Along with p53, another key-role effector of cell cycle homeostasis, pRb, was demonstrated to be affected by embryonic factors treatment via alteration of its phosphorylation state. (Fig. 15)
In vivo results
The effect of embryonic factors on tumor growth was also observed in vivo by s.c. syngeneic injection of primary Lewis lung carcinoma cells into C57BL/6J female mice.7 Immediately before injection, tumor cells were challenged with substances taken from 9 days-pregnant mice uterus, non-pregnant uterus, 9 days-old embryos or liver (the latter being a negative control). Only cells which were mixed with the substances taken from pregnant uterus failed to originate primary tumors: the growth rate was null along the overall experiment time. (Fig. 16) Similar results were obtained in another experiment with the administration of substances taken from Drosophila embryos at the blastodermal stage to mice injected with Lewis lung tumoral cells: 15 days after the inoculation of cells, the primary masses were reduced in the treated animals, with a 35% decrease of tumour weight.8 (Tab. 1)
The evidence that substances taken from pregnant uterus abrogated tumour cell growth, whereas isolated mammals embryo substances did not induce any slowing effect, suggested that factors involved belong to the mother-embryo cross-talk molecular network, and that breaking this interaction abrogates the anti-proliferative effect. Anyway, turning to an oviparous model, it was observed that the embryo alone delivers proliferation-slowing factors, although less efficient than those of pregnant uterine mucosa. It may be that the uterus-embryo system of viviparous animals represents a further evolution of an intrinsic embryonal capability of keeping abnormally proliferating cell clones silent.
|
 |
|
Conclusion
It is apparent from the in vitro and in vivo experimental data that abnormous cell proliferation is somewhat affected by factors which are found in embryos and / or in pregnant uterus. These factors are organized in a network whose complexity should be unscattered to retain its full efficacy. This is particularly true for embryos, whose complex of molecular factors represents a closed microenvironment which is able to normalize the behaviour of abnormously growing cell populations via a regulatory process involving key-role proteins of cell cycle homeostasis. As for molecules which characterize mother-embryo cross-talk, they may mediate a more rapid, apoptosis-enhancer process which does not necessarily need the integrity of the network. In fact, LPF represents a low molecular weight fraction of the whole pig pregnant uterine mucosa, and it does slow tumor cell proliferation rate alone as well as the whole raw homogenate.
A fundamental topic of our findings is that only networks present in differentiative stages of embryo development are able to delay tumour growth, since networks present in multiplicative stages are uneffective or even modulate a slight proliferative effect on cell lines.
Indeed, the future steps of our research will be the isolation of the single components of these networks. The characterization of each single molecule involved in this kind of regulation will be important to deepen in the knowledge of the narrowest mechanisms of how this network works. An intriguing aim will be to focus at the role of stem cells. Stem cells can be committed as well into different cellular types according to the network of factors which constitute the surrounding microenvironment. For example, neural stem cells can be differentiated into cells of the hematopoietic lineage when put in contact with the hematopoietic microenvironment9, or into skeletal muscle cells when put in contact with factors of skeletal muscle differentiation.10 In turn, early-committed stem cells may express a network of differentiating factors which, if administered to a tumour cell, could bring it to normalization. We are studying now the stem cells differentiation factors present during 50% epiboly stage, in which the embryonic totipotent stem cells are differentiating into the adult pluripotent stem cells.
These studies are in progress and they are demonstrating that these factors are able to delay tumor growth, by the control of cell cicle machinery of tumor cells. These studies will be the subject of the next review.
Pier Mario Biava, D. Bonsignorio
Fondazione per la Ricerca delle
Terapie Biologiche del Cancro. Ospedale
di Sesto S. G. - Viale Matteotti 83
Milano, Italia
Janis V. Klavins
Albert Einstein College of Medicine,
New York, NY, USA
